soil surface crust formation : contribution of...

’i ;-

es===+-

737

-

Soil surface crust formation: contribution of micromorphology

L.-NI. Bresson’ and C. Valentin2 lhzsritirr National Agronomique Paris- Grigiioil, 78850 Thivewal-Grignon, France and

Visiting Scientist at CSIRO Division of Soils, Canberra, Australia 20RSTOM. B. P. 11416, Niamey, Niger

ABSTRACT

Bresson, L.-M. and Valentin, C., 1994. Soil surface m s t formation: contribution of micromorphology. In: A.J. Ringrose-Voase and G.S. Humphreys (Editors), Soil Micromorphology: Studies in Management and Genesis. Proc. IX Int. Working Meeting on Soil Minomorphology. Townsville, Australia, July 1992. Developments in Soil Science 22, Elsevier, Amsterdam, pp. 737-762.

Surface crusting is due to the breakdown of surface aggregates into finer fragments and/or primary particles which arc then redistributed on the surface or within the top few millimetres. Microscopic investigations of such thin layers have been found useful for more than ffty years. To assess the specific contribution of micromorphology, a litcrature review of the period 1939- 1991 (54 papers, including only 8 pre-1980) was carried out. The main features of the cxpcriments were analyzed, e.g. soil material propcrtics, soil initial state and rainfall chnrnctcristics. Techniqucs for monitoring crust developmcnt and standards of description and illustration wcre also cxamincd. A limitxion of many experiments was the lack of recognition of crusting stages. Many crusts described seemed to be actually depositional or erosion crusts, which might explain contradictions in the literature about structural crust formation. h i t i d conditions, such as the soil water content before rainfall or the aggregate size distribution, were often not quoted or taken into account in the discussion. The lack of clear description, good illustration and definition of diagnostic features induced misunderstanding of widely used concepts, e.g. “washing-in”. Nevertheless, microscopy has played a major role in our undcrstanding of the various processes involved in both aggregate breakdown and redistribution of the resulting particles. according to soil, climatic and management conditions. In the field, crusts developing during surface degradation are genetically related and form specific time- and spacc-dependent sequences (structural, erosion and depositional crusts). Microscopically-defined crust types can be identified in the field using morphological diagnostic features. Thcse help to assess the crusting rate and allow indentifkation of the processes involved. Methodological recommendations for future micromorphological studies arc made. Research opportunities are also suggested, including the study of: (1) some crusting proccsses, e.g. dispersion in sodic environments and compaction on sandy soils; (2) the complex evolution of crusts for months (cultivated fields) or years (rangelands); (3) the development of crust strength according to the related distribution of coarse and fine particles and (4) the detachability and erodibility of various crusts.

L .,,. i a $ - h irr, kr381us busumeniraif@

, . -- -5 r --

38 <*%

t INTRODUCTION L.-M. BRESSON AND C. VALENTIN

urface (crusting) strongly reduces the infiltration rate, s water storage in the soil and triggers runoff and hence soil erosion. Therefore, and environmental effects are severe, although in arid areas runoff inducement

harvesting management. Moreover, surface crusts cafi induce failure ling emergence and hamper stand establishment. Predicting actual surface degradation

in the field and preventing its consequences requires the knowledge of the processes which take place in a given soil material according to the prevailing management practices and the expected climatic conditions (Bresson and Boiffin, 1990).

Surface crust formation under rainfall or irrigation involves two stages. The first is the wn of surface aggregates into fmer fragments and/or primary particles and involves processes such as slaking due to entrapped air compression and physico-chemical

on of clay. The second is the redistribution of the resulting particles and/or fragments and/or overland flow or verticaUy within the top few

arious types of crusts and/or microlayers have been

ing crusts (Bresson and Boiffm, 1990), erosion pavements (Valentin, ts (Bresson and Cadot, 1992). e name has been used to quali@ different crusts or microlayers, e.g.

er'. Conversely, different names were used to quallfy the same d 'sedimentational', or 'filtration pavement' and 'three-layered

crust'. This has induced some confusion in the literature so that Mualem et al. (1990) t "too much was left undisclosed under the common definition of soil seal leaving too

rpretation". Despite some apparent contradictions, however, tributed to our present understanding of crust formation.

marized the morphological characteristics of surface crusts and fieir genesis in relation to soil, rainfall and topographic characteristics.

The aim of this critical review paper is: (1) to assess the specific contribution of micromorphology; (2) to ge;a better understanding of how the main results could be obtained; and (3) to suggest research opportunities. First, the main features of the experiments were

initial state and rainfall characteristics. In addition,

examined. Then, the main formation processes were discussed and integrated within the framework of the crusting model suggested by Boiffin (1986), Valentin and Ruiz-

RES OF THE REVIEWED PAPERS

paper dealing with soil surface crust and including micromorphological was published in 1939 (Duley). Forty years later, only 8 such papers had been ithin 5 years, between 1986 and 1990,28 micromorphological studies of crusting

SOIL SURFACE CRUST FORMATION ~ 739

contributed to 6 papers and a French one io 5 papers. Four authors, from Australia, France, The Netherlands and Israel, contributed to 4 papers and five others to 3 papers (France, Israel, Italy and USA). Seventeen authors contributed to two papers, and 65 to only one paper.

Objectives

The aim of most papers (83%) was to determine the processes involved in surface crusting. Half of them also dealt with the hydrological and/or mechanical behaviour of crusts. Few studies only considered the behaviour of the crust (9%). Sometimes the main objective of the study was not surface crusting alone but the overall degradation of the cultivated layer, including slumping or compaction (Dexter et al., 1983; Paghi , 1987; Moran et al., 1988; Kooistra et a¿., 1990).

Materials and methods

Soil material properties The particle size distribution of the soil materials studied showed a very wide range: 1.7 -

60.4% clay, 1 - 87.2% silt, 1.2 - 89% sand (the limit between the silt and sand fractions sometimes differed, but this could not be taken into account). First studies dealt with sandy loams and silt loams (Duley, 1939); clay loams andgilty clay loams were studied 30 years later (Evans and Buol, 1968) as well as loams (Ahmad and Roblin, 1971). Then, sandy clay Ioams (Figueira and Stoops, 1983), silty clays (Pagliai et a¿., 1983), clays (Gal er aL, 1984), sands (Valentin, 1986) and silts Ovest et a¿., 1990) were investigated. Valentin (1986) considered the most sandy soil materia1 (89%), West et al. (1990) the most silty (82.2%). The highest clay contents were around 60%, with sand contents ranging from 5% (Moss, 1991) to 30% (Norton, 1987). No study dealt with sandy clays.

The amount of organic matter was given in only 45% of the papers reviewed. The f i s t was Ahmad and Roblin (1971). Organic carbon ranged between 0.05% and 5%.

Gal et al. (1984) were the fxst to consider the ESP of the soil material studied. Usually, ESP was not indicated, which probably means that the soil material studied was not sodic. Therefore, 85% of the soil studied had an ESP < 5, and 15% an ESP > 15 (up to ESP = 88, Greene et a¿., 1988)

In 54% of the papers reviewed, no additional information was given about the soil material. Clay mineralogy or ECC were often given (26%), as well as aggregate stability index (20%).

Different soil materials were studied in only 39% of the papers reviewed. The range of textures greatly varied, but was usually wider for the sand fraction ( > 10% in 95%) than for the clay fraction ( > 10% in 65%). Evans and Bu01 (1968) were the first to study soils with significant differences in clay content.

Sulface characteristics Fifty two percent of the papers reviewed dealt with seedbeds studied in the field; 39% with

soil samples packed in the laboratory and 13% with rangeland soil surfaces (the f i s t being Valentin, 1986). In some cases (7%), different types are compared (the first being Chen et a¿., 1980).

Usually, the aggregate size distribution was not indicated (93%). Ahmad and Roblin (1971) were the first to consider the sorting degree. The upper limit of the aggregate size was quoted

8.'

d ;y v 740 L.-M. BRESSON AND C. VALENTIN

x . . - 8

f L in 39% of the reviewed papers. It ranged between 2 and 40 mm and only seven studies dcalt with aggregates coarser than 4 mm.

The area of the studied plots ranged between 15 and 100,000 cm2. The dope of the soil surface was indicated in 31% of the papers reviewed and ranged between 1.5 and 9%.

The initial moisture content of the soil before rainfall was not quoted in 59% of the reviewed papers, but in most cases a dry initial state could be assumed. Therefore, only one

"third of the studies dealt with moist or wet soils (the first being Tackett and Pearson, 1965). Only five papers compared different initial moisture contents: Valentin (1986), Valentin and Ruiz-Figueroa (1987), Helalia et al. (1988), Le Bissonnais et al. (1989) and Valentin (1991).

Rainfall characteristics Ivlost studies were carried out using simulated rainfall (63%) compared to natural rainfall

(354 ) . Sometimes, flood irrigation was considered (6%). Only 7% of the reviewed papers considered various water application modes: Falayi and Bouma (1973, Cascnave and Valentin (1989), Le Souder et al. (1990) and Bresson and Valentin (1990).

Rainfall intensity, when quoted (56%), was generdly high with an average of 63 mm h-' and a standard deviation of 38. Chen et al. (1980) were the first to use a moderate simulated rainfall intensity (26 mm h-1) and Miicher et al. (1981) a very low intensity (7 mm h-1). The kinetic energy was first quoted by Falayi and Bouma (1975), but few studies did the same (20%). Few studies (15%) considered various intensities like McIntyre and even fewer dealt with various kinetic energies (Valentin and Ruiz-Figueroa, 1987; Levy et al., 1958; Valentin, 1991).

Most studies dealing with simulated rainfall did not specify the characteristics of the water used (92%) and only two considered various electrical conductivities (Tarchitzky et al., 1984; Helalia et al., 1988).

Monitoring crust development Most studies (65%) dealt with only one sampling time, which means that the developmcnt

of the crust could not be studied. The first study which monitored crust development was carried out by Chen et al. (1980). The closest monitoring (16 sampling times) was done by Luk et al. (1990), which allowed them to describe a complex sequence of sub-processes, including four successive generations of vesicles during 30 min rainfall.

The sampling time was generally determined by the duration of the experiment. In most studies dealing with simulated rainfall, this duration was fixed in such a way to get thc so- called "fmal infiltration rate". In field studies under natural rainfall, however, no criteria for the sampling time was given in relation to climatic events or crop development. Time to ponding or runoff was sometimes the criterium for sampling time (9%). In some studies (9%), the development of the crust was monitored using the morphological aspect of the soil surface, according to the method suggested by Boiffin (1986).

Physical characterizations Infdtration rate was assessed in 52% of the papers reviewed, most often from runoff

measurements (68%). Boiffin and Bresson (1987) gave drip infiltrometer measurements. Other physical characteristics are sometimes given, such as the bulk density of the soil surface (13%) or the penetration resistance of the crust (11%). LÆ Bissonnais et al. (1989) studied

SOIL SURFACE CRUST FORMATION 741

crust porosity using mercury injection and the modulus of rupture of thc crusts was measured by Evans and Buol(l968) and Greene et al. (1988).

Microinorphological observations Microscopy was the main tool for 60% of the reviewcd papers. Thin sections of

impregnatcd crusts werc studied in most papers @I%), although Duley (1939) did not observe thin sections but microprofiles of crusts. Scanning electron microscopy of bulk samples was commonly used (41%), following Chen et al. (1980). Both techniques were often used simultaneously (20%), the first being Pagliai et al. (1983).

Magnifications rangcd between xl and ~10 ,000 . Half of the reviewed papers exhibited illustrations or descriptions at different mágnifications and the ratios between the lowest magnification and the highest one ranged from 2 to 2,500.

Descriptions were gcnerally of moderate quality. Mücher et al. (1981) were the first to provide quite full and global descriptions and the first detailed descriptions could be found in Chartres et al. (1985). Illustrations were not always very informative. Crossed or semi-crossed polarized images of thin sections, as well as SEM micrographs, were often disappointing. The photographs given by Duley (1939) were of great quality.

Image analysis was used in 30% of the reviewed papers, mainly for porosity characterization.

Results I

Type of the crust studied Structural crusts were observed in 41% of the reviewed papers. Most were named as such

following Chen et al. (1980) but others were namcd using different terms although the formation processes invoked were typical of structural crusts. In addition, the crusts described and/or illustrated in some studies (20%) also seemed to be structural crusts.

So-callcd "depositional crusts" (Chen et al., 1980), or crusts explicitly corresponding to this concept suggested by Evans and Bu01 (1968), were observed in 44% of the reviewed papers. In addition, crusts described andor illustrated in some studies (20%) also seemed to be depositional crusts.

Erosion crusts (Mücher et al., 1,981) were described in 15% of the papers reviewed, namely by Valentin and Ruiz-Figueroa (1987), Escadafal and Fedoroff (1987), Mücher et al. (1988), Poss et al. (1989). Casenave and Valentin (1989), Kinne11 et al. (1990) and Valentin (1991). In addition, so-called "cryptogamic crusts" were described by Mücher et al. (1988) and by Chartres and Miicher (1989).

Concepts developed Generally, the main conclusions of the reviewed papers dealt with the processes involved in

crust formation (78%). Nevertheless, the concept of typical microhyers often arose (43%). The role of microtopography in crust differentiation was only occasionally considered (22%), namely by Falayi and Bouma (1975), Valentin (1986) and his associates, Boiffm (1986) and his associates, Norton (1987) and Levy et al. (1988). The concept of crusting stages could be found in 32% of the reviewed papers, but this global proportion overestimates the actual use of this essential concept. In fact, the number of authors concerned was few: Tarchitzky et al. (1984), Valentin (1986) and his associates, Boiffin (1986) and his associates, Norton (1987)

Main mechanisms

Mechanisms ~ suggested

L.-M. BRESSON AND C. VALENTIN

Main mechanisms suggested in the reviewed papers: frequency, first quotation and main

Mechanisms Papers First quotation Main references ~ suggested involved

(%Io)

suggested in the


reviewed papers: frequency, first quotation and main

Papers First involved

(%Io)

quotation Main references

t - Slaking 39 McIntyre (1958b)

35 30 12

11 washing-in - 37

39

44

7

McIntyre (1958a) McIntyre (1958a) Ahmad and Roblin (197 1) McIntyre (1958a) McIntyre (1958a) '

Duley (1939)

Evans and Buol (1968) Miicher et al. (1988)

cryptogams 1 . - 2

Tarchitzky et al. (1984); Valentin (1986); Le Bissonnais et al. (1989) Greene et al. (1988); Helalia et al. (1988) Moss (1991) Dexter et al. (1983); Bresson and Boiffin (1990) Mchtyre (1958a); Miicher et al. (1981) Greene et al. (1990); Bresson and Cadot (1992) Norton (1987); Radcliffe et al. (1991); West et al. (1990) Miicher and De Ploey (i977); Miicher et a2. (1988); Bresson and Boiffin (1990) Miicher et al. (1988); Chartres and Miicher (1989); Greene et al. (1990)

uk er al, (1990). Chen et al. (1980) distinguished different stages in structural crust did not suggest any relationship with the depositional cmst. Tarchitzky et al. different cmst morphologies to the hydrological conditions at the soil surface off, after runoff and steady state infiltration rate. The genetic relationships

between structural and depositional crusts were fxst described by Valentin (1981) and Boiffin

Mechanisms suggested The main mechanisms suggested in the reviewed papers are listed in Table 1.

INTEGRATED DISCUSSION

~ - 1 ' Crust formation processes

used by Mchtyre (1958a) to describe "a dense layer 0.1 mm thick pore under high magnification". This skin seal "apparently formed act" (McIntyre, 1958a). Even though no description or illustration might be depleted of clay, because it was opposed to "the 'washed- sity" which could be observed underneath. Many authors used the

of 'skin seal' with reference to Mchtyre but to describe another type of layer, enriched in particles compared to the underlying material. For instance, Chen et al. (1980) described

743 SOIL SURFACE CRUST FORMATION

"a dense 'skin' 0.1 mm thick composed almost solely of fme particles". Norton (1987) and Arshad and Mermut (1988) also described as skin seal a very thin and fine textured layer ob3erved at the surface of some crusts. However, such plasmic seals can result from two different processes: (1) preferential erosion of the coarse particles, which form the top of the crust, by overland flow (Chen et al., 1980; Valentin and Ruiz-Figueroa, 1987) or (2) deposition of clay particles dispersed in turbulent water flow when rain stops (Paghi et al., 1983; Norton, 1987; Arshad and Mermut, 1988).

According to the process involved, i.e. compaction, erosion or deposition, the impact of the crust on the hydrological behaviour of the soil surface might be different.. Compacted seals and erosion plasmic seals seem to have a great effect on the permeability of the soil surface (Mchtyre, 1958; Valentin, 1991). Conversely, 'after rain deposits' (Fagliai e f al., 1983) which form after overland flow stops, may have little effect on the infiltration rate during the rainfall event. This apparent confusion in the literature may explain why Mudem et al. (1990) stated that "no universal seal (such as 'skin' 0.1 mm) can accurately represent a real seal layer". Micromorphology, however, is a unique tool for identifying these different types of microlayers.

Wushed-in layer formation McIntyre (1958a) defined the concept of 'washing3n' as the "plugging of the large pores by

washed-in material". This concept has been so widely adopted that, when no washed-in layer could be observed, this was often noted (Chen et al., 1980; Tarchitzky et al., 1984; Norton et aL, 1986; Boiffm and Bresson, 1987; Mücher et al., 1988). Moreover, analyzing the results of Chen et al. (1980), Mualem et al. (1990) claimed that "to our best judgement, the presence of washed-in material under the 'skin' was evident", despite the fact that the former authors pointed out the lack of such a layer. Sometimes, the presence or absence of a washed-in layer can be explained easily because crusting involves different processes controlled by soil and environmental conditions Bresson and Boiffm, 1990; Valentin and Bresson, 1992). For instance, such an illuviation of fme material can hardly occur on initially dry seedbeds because slaking induces a rapid breakdown of the aggregate framework which seals the surface and prevents any further illuviation of fine particles (Bresson and Cadot, 1992). Difficulties in identifying small amounts of washed-in fine material against the matrix were also invoked to explain that a washed-in layer could not always been recognized (West et al., 1990; Luk et al., 1990).

However, some confusion in the literature arises because the washing-in process is still not clearly defined, especially the size and composition of the washed particles. McIntyre (1958a) used the expression 'fme particles' and not 'clay particles', even though he pointed out the role of physico-chemical dispersion. This may mean that this fine material consisted of fragments'of aggregates as well as primary particles. Unfortunately, McIntyre did not give any information about this. Despite the prominent role of micromorphology in the elaboration of the washing-in concept, the thin section description was laconic: "considerably reduced porosity, to a depth of 1.5 to 3 mm, due to plugging of the larger pores by washed-in material". Moreover, no illustration was presented. However, most later authors, quoting McIntyre clearly considered the washing-in process as clay particle illuviation, which involves different mechanisms and conditions. Rather surprisingly, they did not provide clear evidence of clay illuviation. Bertrand and Sor (1962), using a labeling method, showed that 1% of the clay moved down to 3 cm depth and Helalia et al. (1988) measured significant amounts of clay in percolating water. only


a few convincing illustrations of clay coatings (related to crusting) can be found in the literature (Greene et a l , 1988; Bresson and Cadot, 1992).

Yet, these two different concepts of washing-in involve different mechanisms and conditions. Some clarification is all the more needed because washing-in was invoked in many papers which did not involve micromorphology. Discussing their results, Agassi et al. (1981) explained that, using distilled water, "the dispersed particles are washed into the soil with the infiltrating water, and the pores immediately beneath the surface become clogged". Yet, no observation was made and the only data given were infiltration rates. Later, Levy et al. (1986) gave Agassi et al. (1981) as the frrst reference when summarizing the washing-in process in their introduction.

&out layer formation

ension in the runoff water and canied down the slope". Such coarse textured surface frequently described later, often using the term 'washing-out' (Norton, 1987)

uced by Onofiok and Singer (1984). The mechanism invoked was usually clay a l , 1984); which was supported by the sorted laminae, composed of 10 - 30

forbation and was then eroded. More recently, West et al. (1990) and Radcliffe et al. (1991) *suggested another mechanism. They observed that such layers occurred in slightly depressed areas, which showed that they were actually depositional features. Valentin (1991) described another type of coarse textured upper layer in sandy soils and related it to a sieving process. Therefore, some discernment is required when using so-called washed-out layers as diagnostic features for idenwing the conditions which induced their formation.

In arid sandy soils, Valentin (1986, 1991) described stuctural crusts with a surface layer of loose skeleton grains, overlying a thin plasmic layer (Fig. 1). Using a close time-dependent sequence of sampling, this author showed that textural differentiation mainly results from mechanical winnowing and sieving so that the finer the particles, the deeper they are deposited. Moreover, the downward translocation of clay through the coarse grained top layer can be enhanced by percolating water. Fine particles then accumulate, probably due to entrapped air within the underlying layers (Kooistra and Siderius, 1986). Raindrop impact plays the main role in sieving crust formation, so that mulching is effective in limiting the development of such

soils, seem to be similar to sieving crusts


Fig. 1. Structural crust: Sieving crust on a sandy soil (vertical polished section, BESI mode): loose skeleton grains overlaying a thin plasmic layer (p).

Infilling crust formation Boiffm and Bresson (1987) described in the field a structural crust with net-like infiilings of

bare silt grains (Fig. 2). Further laboratory studies (Le Bissonnais et al,, 1989) showed that such features developed only if the soil was wet before rainfall and that, with air dried sampl&, a slaking crust quickly sealed the surface. Such crusts were clearly due to silt illuviation. Raindrop impact, rather than physico-chemical dispersion, induces textural separation at the top of surface aggregates and the resulting separated silt grains illuviate a few millimetres deeper into the interaggregate packing voids (Bresson and Cadot, 1992). Contrary to slaking crusts, infilling crusts develop slowly. As a result, they occur only when the soil andlor climatic conditions are unfavourable for more rapid processes such as slaking due to entrapped air compression or aggregate coalescence due to plastic deformation.

k. - L.-M. BRESSON AND C. VALENTIN

Fig. 2. Structural crust: InfïUing crust on a loamy soil (vertical thin section, plain light): net-like ini3hgs of bare silt grains (arrow) . within interaggregate packing voids.

one of the breakdown processes invoked by McIntyre (1958b). Ahmad and escribed a crust where the structure of the top 2 mm of soil had completely

to entrapped air compression was well documented by (1950), but few clear micromorphological illustrations of slaking-induced fore L e Bissonnais et al. (1989) and Le Souder et al. (1991). Tarchitzky et laking as caused either by swelling or the pressure of entrapped air, but

compared to raindrop impact. In the early stage of the ion, Onofîok and Singer (1984) described a reduction in the size of the aggregates

ding increase in micropores, which is good evidence of regate breakdown suggested by Valentin (1981) and Farres (1987). Typically,

, dense layer which do not show clear textural separation particles (skeleton) and fm? particles (plasma), even in sodic soils (Valentin


Fig. 3. Structural crust: Slaking crust on a loamy soil (vertical thin section, plane light).

and Bresson, 1992). Slaking crusts predominate when the soil is dry before rainfall (Valentin, 1981; Boiffm, 1986; Norton, 1987; L e Bissonnais et al., 1989). Hydrophobic conditioners are effective in delaying slaking crust formation (Le Souder et al., 1990, 1991), but mulching is not (Valentin and Ruiz-Figueroa, 1987), which is consistent with the process involved.

'Swelling crusts', observed in arid loamy soils (Valentin, 1991), can be considered as a peculiar form of slaking crust (Valentin and Bresson, 1992). Such crusts are characterized by the banded distribution of skeleton grains within superficial parts of clods. Upon wetting, clay lattices expand, tum into a slurry and fill the interstices between clods (Valentin, 1991). The latter process can be related to coalescing, as well as to the early stage of depositional crust formation ('muddy flow'; Bresson and Boiffm, 1990).

1

I

Coalescing crust formation Bresson and Boiffm (1990) described structural crusts which were rather porous and i

1 showed a diffuse boundary with the underlying undisturbed layer (Fig. 4). Macropores were

. . -,i


Fig. 4. Structural crust: Coalescing crust on a loamy soil (vertical thin section, plane light): rather thick crust showing a diffuse boundary with the underlying porous layer; convexities of packing voids gradually' increase towards the surface.

~ typical polyconcave packing voids at the bottom, but their amount and roughness gradually decreased towards the surface, as convexities developed. The main process involved was a gradual coalescence of the initial aggregates by raindro'p compaction under plastic conditions. Coalescing crusts occur on soils which are wet before rainfall and are most developed in sodic soil (lsresson and Boiffm, 1990). Coalescence of aggregates was described by Ahmad and Roblin (1971) below a 2 mm thick very dense surface layer. Moss (1991b) also described crusts where compaction was mainly due to aggregate deformation. However, studying the overall slumping of seedbeds, Dexter ef al. (1983) suggested an intemal soil erosion process with dispersed material from the finei aggregates welding the coarser aggregates at their points of contact. Some of the crusts described in the literature are similar to coalescing crusts (Evans and Buol, 1968; Moss, 1991b). Remley and Bradford (1989) displayed a thin scction micrograph which shows some morphological features of a coalescing crust. However, thcy

-" did not identify any crust, because the matrix did not show any particle size segregation with depth.

c

1 !

I

1

!

1

I

I

I

!


Silt-lqer indiiced compaction Compaction by raindrop impact was invoked by McIntyrc (1958a), but was not well

documcntcd in later micromorphological studies. Thirty three years later, Moss (1991a and b) provided new concepts. Studying crust formation under simulated rainfall, Moss (1991a) carried out a very close micromorphological monitoring of the process and sampled the early stage after 30 s (0.33 mm rainfall). Craters developed first, then, at 1 min, a thin layer of tightly packed silt grains spread on the surface. After 8 min, the silt layer, 200 pm to 3000 pm thick, covered the surface and compaction was visible to 3 - 5 mm depth. Further experiments involving various kinetic encrgies as well as local shielding of thc surface and telescopic obscrvations, suggested that the silt layer played the major role in compaction of so-called 'rain impact' crusts (Moss, 1991a and b). Three stages could be distinguished: (1) 10 - 50 pm particlcs are concentrated at the surface by preferential removal of other sizes in the airsplashing environment; (2) the resulting silt grains are spread over the surface by lateral outflow sheets of the drops and deposit as tightly packed bed-load sediments; (3) this layer is dilatant, resists deformation by raindrop impact and prevents water penetration because its pores are < 15 pm. Therefore, the underlying layer may be compacted by stress waves (Moss, 1991a and b).

Such a silt layer may be similar to the washed-out layer described in the literature or to the sieving crusts observed on arid sandy soils by Valentin (1986, 1991) and Poss et AL. (1989). However, it has never bcen observed on temperate cultivated loamy soils in France. Contrary to Moss's assertion, silt-layer compaction does not seem to bc "widespread". Moss's experimcnts dcalt with pre-wetted soils, which might prevent slaking, and with poorly aggregated soil materials, which might prevent silt infilling and could not involve aggregate coalescence. Besides, using a wider range of soil materials, Moss (1991b) also described plastic deformation of aggregates and washing-in of fine particles. Therefore, the innovative conception of compaction suggested by Moss seems to be compatible with the model of structural crust formation suggested by Bresson and Valentin (1990). Further studies are required for dctermining the conditions which lead to silt layer induced compaction, especially in the field.

Erosion crust forinatiori Erosion crusts were defined by Valentin (1981, 1991) as thin, smooth surface layers

enriched in fine particles (Fig. 5). The fine particles are usually poorly oriented. Voids are generally restricted to some cracks and vesicles. The thickness of this plasmic layer is rather regular and is not related to the surface microtopography. Some skin seals described in the literature may be similar (Chen et d., 1980). Kinne11 et al. (1990) observed that, after erosion of the first top millimetres, the remnants were mainly argillaceous layers. Such crusts often result from erosion of thc coarse textured top layer of sieving crusts (Valentin, 1986). They form a rather resistant surface against further wind or water erosion, and therefore often cover large areas. Erosion crusts form first on the higher points, then expand over the surface as the global surface roughness diminishes. Formation of an erosion crust from other structural crusts, which do not show any textural differentiation, may involve the preferential erosion of coarser particles by high energy raindrops (Chen et al., 1980; Valentin, 1991). Erosion crusts can be recognized from after rain deposits and eroded depositional crusts using not only their spatial distribution in the ficld but also micromorphological features of their plasmic surface layer, namely (1) the poor orientation of the fine particles and (2) the absence of relationship

J' y 750 L.-M. BRESSON AND C. VALENTIN

Fig. 5 . Erosion CI

sandy soil (vertical section, BESI smooth plasmic layer (arrow).

ust on a polished

mode): surface

between the layer thickness and the surface microtopography. However, the term 'erosion crust' might be confusing because most crusts affected by erosion processes are not 'erosion crusts'.

arid rangeland areas rather than in cultivated fields, presumably because the velocity of overland flow is not limited by the surface roughness of the seedbed. Moreover, rangeland crusts are not rejuvenated by tillage practices and often develop over many years (Valentin and Bresson, 1992).

Erosion crusts are usually more developed

Depositional crust formation Crusts formed by deposition of the particles suspended in overland flow (Fig. 6) were

recognized by Evans and Buol (1968). The term 'depositional crust' was later introduced by Chen et al. (1980). The micro-sedimentation process involved (Bishay and Stoops, 1975) was studied by Miicher and De Ploey (1977). Sedimentology provided the basic concepts which were transposed to the microenvironment of crust formation. Micromorphology was found a

I

i

! i

I

I

I i

i

i !

1

i !

SOIL SURFACE CRUST FORMATION 75 1

Fig. 6. Depositional crust on a loamy soil (vertical thin section, plane light): microbedded layers overlying a structural crust.

very useful tool for determining the main diagnostic characters, i.e. microbedding, particle sorting and orientation.

Sortiizg of basic particles was studied by Miicher et al. (1981) in both erosional and depositional environments. Deposits resulting from turbulent rainwash generally lead to laminae with 10 - 50 pm grains. Conversely, afterflow deposits show laminae with a greater percentage of particles smaller than 30 pn-. Pure splash deposits do not show any lamination or particle sorting (Miicher et al., 1981). The presence of aggregates within depositional crusts was also described (Falayi and Bouma, 1975). Miicher et al. (1988) observed loose aggregates at the top of crusts, and suggested they were deposited by aftefflow. Conversely, microbeds observed at the bottom of depositional crusts, filling in small pocket-like depressions in former structural crusts, contained small aggregates included within a densely packed material. This was related to a muddy flow-process which induced short-distance translocation of particles (Bresson and Boiffin, 1990).


entation of the deposited particles is related to the degree of dispersion. Studying the of electrolyte concentration on the micromorphology of artificial depositional crusts, d et al. (1988) showed that only suspensions prepared with distilled water produced

s distribution pattems of microbeds have been described, e.g. 'rill' and 'figer-like' and Siderius, 1986; Mücher et al., 1988) and 'deltaic' (Bresson and Boiffm, 1990).

uperposition of different pattems has been used to assess the succession of different hydrodynamic conditions at the soil surface during crust development (Bresson and Boxfin, 1990). Moreover, studying a long-term experimental field to which various fertilizers and amendments had been applied, Bresson and Boiffin (1990) suggested that the size and duration of puddles played the main role in the characteristics of the depositional crust which, in tum, appeared to be partly controlled by the properties of the underlying structural crust. This

crusts with high birefrin, aence.

lied to all the studied soils, including a sodic soil.

togamic crusts have been recognized for many years in rangeland areas (Fletcher and 1948). However, the first microscopic characterization of such crusts were carried out

ecently in Australia (Mücher et al., 1988; Chartres and Mücher, 1989; Greene et al., 1990). Cryptogams develop preferentially on argillaceous materials

and Ca-oxalate crystals exuded from plant roots during stable periods erosion &Eicher er al., 1988). Amorphous gel-like organic material has

ed with algal sheaths, and sometimes fungal hyphae extend a w millimetres below surface cryptogams (Greene et al., 1990). Individual algae, which often

osition with lichens, contribute to aggregate stabilization by secreting cementing and Harris, 1964), reinforcing the aggregation effect of associated fungal hyphaes

The effect of cryptogamic crusts on the hydrological behaviour of the surface, however, is Il understood, and may be either beneficial or detrimental (van der Watt and Claassens, Generally, such crusts are considered as a good protection against erosion due to their cohesion Fletcher and Martin, 1948; Greene and Tongway, 1989). Destruction of the gam cover by fire gives very clear evidence of its protective effect (Chartres and r, 1989; Greene et aL, 1990; %ell et al., 1990). However, Greene et al. (1990) found

runoff could be greater from cryptogam mats than from bare surfaces and removal of the togamic crust 'increased infiltration rate four times (Greene et al., 1990). This may be

d by the properties of the underlying layers which may control infiltration rate. amic cover were observed on various types of crusts, erosion crusts and depositional

n and Bresson, 1992). Therefore, cryptogams should be considered as a micro rather than a micro soil-layer, and thus should not be studied alone but in

. . ,rebtion with the crust they colonize. ~. 1 .

Ve&les have been described in various crust types and in various environments. Springer sted that vesicles form by air entrapment. When the saturated crust dries out it so that escaping gases form cavities. The production of carbon dioxide liberated

from organic material may also play a role in vesicle formation (Pagliai et and Stoops (19S3) described particular vesicles which showed micro-

- .


erosion of the walls and micro-deposition on the bottom by the action of a descending water front. Usually, vesicles are more numerous and larger as the number of wetting and drying cycles increases. Miller (1971) showed that wetting and drying cycles first induced a platy strusture in which vesicles then gradually form due to capillary pressure on wetting. However, Figueira and Stoops (1983) suggested that the platy structure is mainly determined by fissures interconnecting the large vesicles previously formed. On the other hand, Bresson and Boiffm (1990) suggested that the genesis of vesicles.may be ascribed to compaction of initial packing voids due to plastic strain under semi-liquid conditions. Although vesicle formation should be better documented, vesicle microlayers can be successfully used as a predicting criterion of very low infiltrability (Casenave and Valentin, 1989).

Crusting nwdel

Valentin (1981) and Boiffm (1984) suggested essential concepts for crusting studies, which were adopted and developed in the following papers: Boiffm, (1986), Valentin (1986), Boiffm and Bresson (1987), Valentin and Ruiz-Figueroa (1987), Le Bissonnais et al. (1989), Poss et al. (1989), Casenave and Valentin (1989), Le Souder et al. (1990), Bresson and Boiffm (1990), Bresson and Valentin (1990), Le Souder et al. (1991), Valentin (1991), Bresson and Cadot (19921, Casenave and Valentin (1992) and Valentin and Bresson (1992). According to these authors, the main different types of crusts are genetically related, and form time- and space-dependent sequences.

Time- and space-dependant variations Time- and space-dependent variability were mainly studied by Valentin (1986) and his

associates and by Boiffin (1986) and his associates. Chen et al. (1980), Norton (1987) and Luk et al. (1990) distinguished different stages in structural crust formation, but did not suggest any relationship to depositional crusts. Tarchitzky et al. (1984) related different crust morphologies to the hydrological conditions at the soil surface, i.e. before runoff, after runoff and steady state infiltration rate. The genetic relationships between structural and depositional crusts were first described by Valenti (1981, 1986) and Boiffm (1984, 1986). The role of microtopography in crust differentiation was seldom considered by other authors. Falayi and Bouma (1975) were the first, followed by Norton (1987) and Levy et al. (1988).

Crusting stages. Crusting is a dynamic process which follows the same general pattern: (1) sealing of the surface by a structural crust, then (2) development of a depositional crust (Fig. 7). The change from the first to the second stage mainly depends on a decrease in infïitration rate due to the structural crust properties, which induces microrunoff (Valentin, 1981; Boiffm, 1984, 1986). These two stages can be identified in the field, using simple macroscopic features (Boiffin, 1984; Bresson and Boiffin, 1990) which can also be used to quantify the crustin, 0 rate (Boiffm, 1986).

Spatial variability. Structural crusts generally develop faster where aggregates are fmer, which explains why crusts observed in the field do not uniformly cover the seedbed surface (Boiffin, 1984, Boiffin and Bresson, 1987). Depositional crusts first form in microdepressions or interstices between large clods. As the soil surface flattens, the deposited microbeds become thinner but tend to expand more extensively over the former structural crust (Bresson and Boiffin, 1990). In sandy soils, the surface roughness is usually more transitory and the spatial variability of crusts occurs at a larger scale: structural crusts being observed upslope, erosion

54

Seedbed Structurnl

L.-M. BRESSON AND C. VALENTM

Depositional crust crust

* sealing

.7. Time- and space-dependent sequence of crusts in a loamy cultivated field: (1) sealing of soil surface by a structural crust, then (2) development of a depositional crust.

i

crusts midslope and depositional crusts downslope (Valentin, 1991; Valentin and Bresson, .'1992).

their studies of loamy and sandy soils, which involved more than 400 thin sections re than 100 soils, Valentin and Bresson (1992) suggested a typology of crusts based

icro-morphological characterization: (1) structural crusts including slaking g, coalescing and sieving (and coarse pavement) sub-types; (2) erosion

3) depositional crusts, with two sub-types, runoff- and still-. Such a morpho- ification of crusts appeared to be relevant to the prediction of infïïtrability (j3oiffm

on and Boiffin, 1990; Casenave and Valentin, 1992; Valentin and

logy seems to account for most of the crusts described in the literature except the clay depleted layer). Skin seal can be related to erosion crusts, depositional crust

their morphology. Washed-in layers, if well identifed, may be part out layers can be related to silt layers or depositional crusts or may

togamic crusts can be considered as crusts of whatever type (according underlying layers) colonized by a peculiar vegetal cover.

- - _. CONCLUSIONS , I

"The long-term benefit of applied research, even if recognized, was sacrificed to meet the ort term 'practical' objective. Consequently, soil sealing is still understood in concepts

cIntyre, '1958" (Mualem et aL, 1990). The current critical review of 54 papers oscopic investigations of soil crusts shows that the latter statement is far fetched. on of micromorphology to our present understanding of crust formation has

en prominent. Various crusting processes have been clearly identifed or elucidated thanks to croscopic observations, e.g. depositional, coalescing, infilling, sieving and slaking crust rmation and silt-layer compaction, Diagnostic macro- and microscopic features have been

ch allow recognition of the processes involved in the formation of the various types (Valentin and Bresson, 1992). Moreover, even though Mualem et al. (1990)

d that "there are serious doubts about the validity of the identification of the seal ed during rainfall, with the crust layer visually observed aftenvard", these

c ~ p i ~ a l l y defmed crust types helped to assess the crusting rate as well as to predict

1


infltrability (Boiffm and Monnier, 1986; Bresson and Boiffin, 1990; Casenave and Valentin, 1992). Micromorphology also helped to assess the role of initial moisture content and initial aggregate size distribution in controlling the nature and the kinetics of crusts. Moreover, the general pattern of soil surface crust development was partly elaborated thanks to microscopic studies. Therefore, micromorphology appears to be a unique tool for predicting and controlling cmst development.

Some suggestions for further studies arise from the current critical review.

Methodological suggestions

Soil initial state and rainfall event characteristics In most studies (87%), the initial moisture coiitent before rainfall was neither specified, nor

taken into account in the discussion. The same situation generally prevailed (93%) for the initial aggregate size distribiction. These two factors, however, partly control the crusting rate as well as the type and the properties of the crust which forms (Falayi and Bouma, 1975; Boiffm, 1986; Valentin and Ruiz-Figueroa, 1987; L e Bissonnais, 1988; Bresson and Cadot, 1992). Underestimating their role may have led to apparent contradictions in the literature. Moreover, controlling the initial state of the seedbed is a realistic method for farmers to delay structural crust formation especially when slaking by air entrapment compression is the main soil degradation process.

In most field studies under natural rainfall (87%), the crust samples studied were not referred to the climatic events which induced their formation. Such data, however, are required for relating the morphological features observed to the physical conditions of their formation, e.g. rainfall intensity, cumulative kinetic energy, dry periods, etc. (Boiffii, 1984, 1986; Bresson and Boiffin, 1990).

Dynamics of crust development Close monitoring. Many studies (64%) dealt with only one sampling time. A closer

monitoring is required to take into account the dynamic aspect of crust formation. This is of great practical importance because the impact of crusting on seedling emergence is closely related to the development rate of the crust compared to the emergence and establishment rate of the crop.

Reference to the crusting srage. In most studies (72%), crust samples were not referred to the development stage they characterized. However, such a reference is needed for reliable comparison of crusting in soils of different composition, climatic environment and land use and management (Bresson and Boiffin, 1990). This is especially important when a close monitoring cannot be done.

Specijïc problems of simulated environments Simrilated seedbeds. Usually, crusts were sampled after runoff had started, i.e. after the

formation of the structural crust. Because the surface of repacked seedbeds was usually very smooth and because a slope was generally set up to prevent ponding, such crusts actually formed under constant erosional conditions. Yet, such experiments were used not only to study erosional processes but also to discuss the formation of structural crusts. On the other hand, most of the crusts formed in the laboratory from repacked samples did not actually seem to correspond to any of the natural crusts described in the field. There is a very close relationship


tween the so-called washed-out layers and this type of experiment. Among the papers layers, 84% dealt with repacked seedbeds and, among those dealing

seedbeds, 76% described a washed-out layer. Among the five papers which did t layers, two seemed to have stopped the experiment before runoff 1; Le Bissonnais et al., 1989). McIntyre did not describe any washed-

but his thin sections, which he did not examine himself (McIntyre, 1958a) and which Soils in Canberra, clearly show similar coarse textured top

s suggests that more realistic simulation of the conditions prevail in the field should be considered, including (1) aggregate size distribution and (2) opography. Moreover, formation of structural crusts should be only studied using crusts

cent studies showed that vzrying intensity rains should also be used. bserved that the hydraulic properties of the seal were mainly

y the rainfall intensities prior to ponding and during intensity peaks. They partly controlled the balance between crust formation and

, -

Rheology ding to many scientists studying soil surface crusting, it is evident that crust formation

ociated with clay dispersion and movement in the soil (Agassi er aL, 1981). Describing a micrograph which seems to be a coalescing crust, Remley and Bradford (1989) did any crust, because the matrix did not show any particle size segregation with

is an example of the close conceptual relationship behveen crusting and dispersion. r, on a sodic soil, Bresson and Boiffin (1990) could not fmd any textural separation in

rusts. Conversely, in a non-sodic environment, clear textural separation occurred, ains illuviating just below the surface aggregates and clay particles depositing a few

deeper. This was related to the remaining aggregate framework (Bresson and 2). El Morsy et al: (1991) also suggested that conditions leading to dispersion rapid crust formation which reduced the potential for illuviation. Therefore, textural should not be related to dispersive conditions. Moreover, the current critical review

in many cases other mechanisms control the nature and the development rate of on sodic soils, swelling and deformation under plastic conditions could play the resson and Boiffm, 1990). Gypsum reduced crust formation more efficiently on

smectitic rather than kaolinitic clays (Chartres et al., 1985; Greene et al., 1988), that swelling occurred rather than real dispersion. of fact, physico-chemical dispersion, swelling and deformation under plastic ot constitute different processes but different levels of the same process, i.e.

st development in clayey and sodic environments should be related to logical properties rather than to the physico-chemical dispersability sensu srricto.

ewed studies dealt with the development of crust during the few weeks was pertinent to the impact of crusting on seedling emergence. However, crust, after harvesting, greatly controls the ability of the soil to store water

I I !

i

!

!

!

i

!

i

1

I

i I I


unti the next sowing. Moreover, this final state may also control the tillage conditions and hence the properties of the next seedbed. In this respect, the mid-term (several months) evolution of crusts should be considered, which requires taking account of the succession of climatic events. Field scale studies are required to predict runoff and water intake. Nevertheless, simulated environments which deal with erosional conditions could also be relevant.

Natural crusts in rangeland areas involved the same processes as crusts developing in cultivated soils. However, they result from a much longer evolution because they are not rejuvenated by tillage practices. Moreover, velocity of overland flow in rangelands is not limited by the surface roughness induced by tillage (Valentin and Bresson, 1992). As a result, rangeland crusts involve a complex succession of erosion, sedimentation and bioturbation periods (Valentin, 1986; Miicher et al., 1988; Casenave and Valentin, 1989). Micromorphology should help to recognize such different phases (Courty, 1986), and to identify specific processes, if any.

c;

Mechanical properties Few micromorphological studies dealt with the mechanical properties of crusts, and the

applicability of the typology suggested in this paper to the prediction of mechanical impedance is yet to be assessed. Micromorphology should be especially useful to study the relationships between mechanical properties and the relative distribution of the coarse and fine particles.

Crust erodibility Crusts greatly control runoff and soil loss. In many studies, the morphology of crusts was

related to their hydrological behaviour, i.e. infiltration rate and runoff inducement. On the contrary, very litde is known about the relationship between detachability and the type and the morphology of crusts. Applicability of the typology presented in this paper should be tested and new research on crust detachability and erodibility should be carried out involving microscopic investigations.

XEFERENCES

* Papers including microscopic study and reviewed here. "Ahmad, N. and Roblin, A.J., 1971. Crusting of River Estate soil, Trinidad, and its effect on

gaseous diffusion, percolation, and seedling emergence. J. Soil Sci., 22: 23-31. Agassi, M., Shainberg, I. and Morin, J., 1981. Effect of electrolyte concentration and soil

sodicity on infiltration rate and crust formation. Soil Sci. Soc. Am. J., 45: 848-851. *Arshad, M.A. and Mermut, A.R., 1988. Micromorphological and physico-chemical

characteristics of soil crust types in northwestem Alberta, Canada. Soil Sci. Soc. Am. J., 52:

Bertrand, A.R. and Sor, K., 1962. The effects of rainfall intensity on soil structure and migration of colloidal material in soils. Soil Sci. Soc. Am. J., 26: 297-300.

Boiffm, J., 1984. La dégradation structurale des couches superficielles du sol sous I'action des pluies. Ph.D. Thesis, Institut National Agronomique Paris-Grignon, Paris, 320 pp.

Boifin, J., 1986. Stages and time-dependency of soil crusting in situ. In: F. Callebaut, D. Gabriels and M. De Boodt (Editors), Assessment of Soil Surface Sealing and Crusting. Flanders Research Center for Soil Erosion and Soil Conservation, Ghent, pp. 91-98.

724-7 29.


I, and Bresson, L.-M., 1987. Dynamique de formation des croûtes superficielles: de l'analyse microscopique, In: N. Fedoroff, L.M. Bresson and M.A. Coutry

rs), Soil Micromorphology. Proc. VI1 Intl Working Meeting of Soil mmorphology, Paris, July 1985. Association Française pour 1'Etude du Sol, Plaisir,

and Monnier, G., 1986. Infiltration rate as affected by soil surface crusting caused by In: F. Callebaut, D. Gabriels and M. De Boodt (Editors), Assessment of Soil Sealing and Crusting. Fianders Research Center for Soil Erosion and Soil

ation, Ghent, pp. 210-217. J.R., 1964. The influence of the microflora on physical properties of iated with filamentous algae and fungi. Aust. J. Soil Res., 2: 11 1-122.

Boiffi, J., 1990. Morphological characterization of soil crust ment stages on an experimental field. Geoderma, 47: 301-325. .-M. and.Cadot, L., 1992. Illuviation and structural crust formation on loamy

L.-M. and Valentin, C., 1990. Comparative micromorphological study of soil in temperate and arid environments. Trans. 14th Congress of I.S.S.S., Kyoto,

. and Stoops,-G., 1975. Micromorphology of irrigation crusts formed on a soil of the mechanized farm, north-west Egypt. Pédologie, 25: 143.

C., 1989. Les Etats de Surface de la Zone Sahélienne. M u e n c e ation. ORSTOM, Paris, France, 230 pp. . and Valentin, C., 1992. A runoff capability classification system based on surface

s criteria in semi-arid areas of West Africa. J. Hydrol., 130: 231-249. , CJ. and Miicher, H.J., 1989. The effects of fire on the surface properties and seed ation in two shallow monoliths from a rangeland soil subjected to simulated raindrop

CJ., Greene, R.S., Ford, G.W. and Rengasamy, P., 1985. The effect of gypsum on orosity and crusting of two red duplex soils. Aust. J. Soil Res., 23: 467-479. , Tarchitzky, J., Brouwer, J., Morin, J. and Banin, A., 1980. Scanning electron

microscope observations on soil crusts and their formation. Soil Sci., 130: 49-55. J. and Valentin, C., 1985. Evaluation of factors influencing water erosion in West

g rainfall simulation. In: Challenges in African Hydrology and Water Resources,

., Smillie, G.W. and Hussain, S.M., 1986. Laboratory studies of crust development and Iraqi soils. III. Micromorphological observations of artificially-formed crusts. ge Res., 6: 337-350. . -A., 1986. Morphology and genesis of soil surface crusts in semi-arid conditions

Boodt ,+Assessment of Soil Surface Sealing and Crusting. Flanders Research Center for ion and Soil Conservation; Ghent, pp. 32-39. R., Radke, J.K. and Hewitt, J.S., 1983. Structure of a tilled soil as influenced by

wheat cropping, and rainfall. Soil Sci. Soc. Am. J., 47: 570-575. L., 1939. Surface factors affecting the rate of intake of water by soils. Soil Sci. Soc.

ate soil. Soil Sci. Soc. Am. J., 56: 1565-1570.

Earth Surf. Proc. Landforms, 14: 407-417.

144, pp. 451-461..

gion, north-west India). In: F. Callebaut, D. Gabriels and M. De

SOIL SURFACE CRUST FORMATION . 759

EI Morsy, E.A., Malik, M. and Letey, J., 199 1. Interactions behveen water quality and polymer treatment on infiltration rate and clay migration. Soil Technology, 4: 221-231.

*Escadafal, R. and Fedoroff, N., 1987. Apport de la micromorphologie à une étude multi- scalaire de la surface des sols en région aride (Tunisie méridionale). In: N. Fedoroff, L.M. Bresson and M.A. Coutry (Editors), Soil Micromorphology. Proc. VI1 Int. Working Meeting of Soil Micromorphology, Paris, July 1985. Association Française pour 1'Etude du Sol, Plaisir, France, pp. 409-413.

*Evans, D.D. and Buol, S.W., 1968. Micromorphological study of soil crusts. Soil Sci. Soc. Am. Proc., 32: 19-22.

*FalaYi, O. and Bouma, J., 1975. Relationships between the hydraulic conductance of surface crusts and soil management in a Typic Hapludalf. Soil Sci. Soc. Am. Proc., 39: 957-963.

Fanes, P.J., 1987. The dynamics of rainsplash erosion and the role of soil aggregate stability. Catena, 14 119-130.

*Figueira, H. and Stoops, G., 1983. Application of micromorphometric techniques to the experimental study of vesicular layer formation. Pédologie, 33: 77-89.

Fletcher, J.E. and Martin, W.P., 1948. Some effects of algae and molds on the rain crust of- desert soils. Ecology, 29: 95-100.

*Gal, M., Arcan, L., Shainberg, I. and Keren, R., 1984, Effect of exchangeable sodium and phosphogypsum on crust structure - scanning electron microscope observations. Soil Sci. Soc. Am. J., 48: 872-878.

Giménez, D., Dirksen, C., Miedema, R., Eppink, L.A.A.J. and Schoonderbeek, D., 1992. Surface sealing and hydraulic conductances under varying-intensity rains. Soil Sci. Soc. Am.

Greene, R.S.B. and Tongway, D.J., 1989. The significance of (surface) physical and chemical properties in determining soil surface condition of red-earths in rangelands. Aust. J. Soil Res., 27: 213-225.

*Greene, R.S.B., Rengasamy, P., Ford, G.W., Chartres, C.J. and Millar, J.J., 1988. The effect of sodium and calcium on physical properties and micromorphology of two red-brown earth soils. J. Soil Sci., 39: 639-648.

"Greene, R.S.B., Chartres, C.J. and Hodgkinson, K.C., 1990. The effects of fire on the soil in a degraded semi-arid woodland. I. Cryptogam cover and physical and micromorphological properties. Aust. J. Soil Res., 28: 755-777.

"Helalia A.M., Letey, J. and Graham, R.C., 1988. Cmst formation and clay migration effects on infiltration rate. Soil Sci. Soc. Am. J., 52: 251-255.

*Kinnell, P.I.A., Chartres, C.J. and Watson, C.L., 1990. The effects of f i e on the soil in a degraded semi-arid woodland. II. Susceptibility of the'soil to erosion by shallow rain- impacted flow. Aust. J. Soil Res., 28: 779-794.

*Kooistra, M.J. and Siderius, W., 1986. Micromorphological aspects of crust formation in a savanna climate under rainfed subsistence agriculture. In: F. Callebaut, D. Gabriels and M. De Boodt (Editors), Assessment of Soil Surface Sealing and Crusting. Flanders Research Center for Soil Erosion and Soil Conservation, Ghent, pp. 9-17.

*Kooistra, M.J., Juo, A.S.R. and Schoonderbeek, D., 1990. Soil degradation in cultivated alfisols under different management systems in southwestern Nigeria. In: L.A. Douglas (Editor), Soil Micromorphology: A Basic and Applied Science. Proc. VI11 Int. Working Meeting on Sail Mircromorphology, San Antonio, Texas, July 1988. Developments in Soil Science 19, Elsevier, Amsterdam, pp. 61-69.

J

I

I

J., 56: 234-242.

J

I

60 L.-M. BRESSON AND C. VALENTIN

Le Bissonnais, Y., 1988. Analyse des Mécanismes de Désagrégation et de la Mobilisation des Particules de Terre sous l'Action des Pluies. Ph.D. Thesis, Université d'Orléans, Orléans, 2 2 5 ~ 1 .

*Le Bissonnais, Y., Bruand, A. and Jamagne, M., 1989. Laboratory experimental study of soil crusting: Relation between aggregate breakdown mechanisms and crust structure. Catena 16: 377-392.

. *Le Souder. C.. Le Bissonnais, Y., Robert, M. and Bresson, L.-M., 1990. Prevention of crust with a mineral conditioner. In: L.A. Douglas (Editor), Soil Micromorphology: A Applied Science. Proc. VI11 Int. Working Meeting on Soil Mircromorphology, io, Texas, July 1988. Developments in Soil Science 19, Elsevier, Amsterdam, pp.

*Le Souder, C., LeBissonnais, Y. and Robert, M., 1991. Influence of a mineral conditioner on

Levy, G., Shainberg, I. and Morin, J., 1986. Factors affecting the stability of soil crusts h storms. Soil Sci. Soc. Am. J., 50: 196-201. , Berliner, P.R., Du Plessis, H.M. and Van der Watt, H.v.H., 1988. graphical characteristics of artificially formed crusts. Soil Sci. Soc. Am. J., 5 2

., Dubbin, W.E. and Mermut, A.R., 1990. Fabric analysis of surface crusts developed under simulated rainfall on loess soils, China. In: R.B. Bryan (Editor), Soil Erosion - Experiments and Models. Catena Verlag, Cremlingen, Germany, pp. 29-40.

McIntyre, D.S., 1958a. Permeability measurements of soil crusts formed by raindrop impact. Soil Sci., 85: 185-189.

Mchtyre, D.S., 1958b. Soil splash and the formation of a surface crusts by raindrop impact. Soil Sci., 85: 261-266.

*Miller, D.E., 1971. Formation of vesicular structure in soil. Soil Sci. Soc. Am. Proc., 35: 635-

*Moran, C.J., Koppi, A.J., Murphy, B.W. and McBratney, A.B., 1988. Comparison of the macropore structure of a sandy loam surface soil horizon subjected to two tillage treatments. Soil Use Manage., 4 96-102.

*Moss, A.J., 1991a. Rain-impact soil crust. I. Formation on a granite-derived soil. Aust. J. Soil Res., 29: 271-289.

*Moss, A.J., 1991b. Rain-impact soil crust. II. Some effects of surface-slope, drop-size and soil variation. Aust. J. Soil Res., 29: 29 1-309.

Mualem, Y., Assouline, S. and Rohdenburg, H:, 1990. Rainfall induced soil seal. A. A critical -'review of observations and models. Catena, 17: 185-203.

*Micher, H.J. and De Ploey, J., 1977. Experimental and micromorphological investigation of ersion and redeposition of loess by water. Earth Surf. Proc., 2: 117-

*Mücher, H.J., De Ploey, J. and Savat, J., 1981. Response of loess material to simulated translocation by water: micromorphological observation. Earth Surf. Proc. Landforms, 6

*Mücher, H.J., Chartres, C.J., Tongway, D.J. and Greene, R.S.B., 1988. Micromorphology and.significance of the surface crusts of soils in rangelands near Cobar, Australia. Geoderma, 42: 227-244.

the mechanisms of disaggregation and sealing of a soil surface. Soil Sci., 152: 395-402.

SOIL SURFACE CRUST FORMATION 76 1

*Norton, L.D., 1987. Micromorphological study of surface seals developed under simulated rainfall. Geoderma, 40: 127-140.

~ *Norton, L.D., Schroeder, S.L. and Moldenhauer, W.C., 1986. Differences in surface crusting and soil loss as affected by tillage methods. In: F. Callebaut, D. Gabriels and M. De Boodt ' (Editors), Assessment of Soil Surface Sealing and Crusting. Flanders Research Center for Soil Erosion and Soil Conservation, Ghent, pp. 64-71.

*Onofiok, O. and Singer, M.J., 1984 Scanning electron microscope studies of surface crusts formed by simulated rainfall. Soil Sci. Soc. Am. J., 48: 1137-1143.

"Pagliai, M., 1987. Effects of different management practices on soil structure and surface crusting. In: N. Fedoroff, L.M. Bresson and M.A. Coutry (Editors), Soil Micromorphology.

Française pour 1'Etude du Sol, Plaisir, France, pp. 415-421. *Pagliai, M., Bisdom, E.B.A. and Ledin, S., 1983. Changes in surface structure (crusting) after

application of sewage sludge and pig sluny to cultivated agricultural soils in northern Italy. Geoderma, 30: 35-53.

*Pagliai, M., La Marca, M. and Lucamente, G., 1983. Micromorphological investigation of the effect of sewage sludges applied to soil. In: P. Bullock and C.P. Murphy (Editors), Soil Micromorphology, volume 1: Techniques and Applications, Proc. VI Int. Working Meeting on Soil Micromorphology, London, July 198 1, AB Academic Publishers, Berkhamsted,

*Pos , R., Pleuvret, C. and Saragoni, H., 1989. Influence des réorganisations superficielles sur l'infiltration dans les terres de Barre (Togo méridional). Cahiers de I'ORSTOM, série. Pédologie, XXV: 405-415.

coastal plains loamy sands. Soil Sci. Soc. Am. J., 55: 223-227.

erosion parameters. Soil Sci. Soc. Am. J., 53: 1215-1221.

I

I Proc. VI1 Int. Working Meeting of Soil Micromorphology, Paris, July 1985. Association

U.K., pp. 219-225.

I *Radcliffe, D.E., West, L.T., Hubbard, R.K. and Asmussen, L.E., 1991. Surface sealing in

*Remley, P.A. and Bradford, J.M., 1989. Relationship of soil crust morphology to inter-rill

Robinson, D.O. and Page, J.B., 1950. Soil aggregate stability. Soil Sci. Soc. Am. Proc., 14:

%outhard, R.J., Shainberg, I. and Singer, M.J., 1988. Influence of electrolyte concentration on the micromorphology of artificial depositional crust. Soil Sci., 145: 278-288.

Springer, M.E., 1958. Desert pavement and vesicular layer of some soils of the desert of the Lahonton Basin, Nevada. Soil Sci. Soc. Am. Proc., 22: 63-66.

*Tackett, J.L. and Pearson, R.W., 1965. Some sharacteristics of soil crusts formed by simulated rainfall. Soil Sci., 99: 407-413.

*Tarchitzky, J., Banin, A., Morin, J. and Chen, Y., 1984. Nature, formation aid effects of soil crusts formed by water drop impact. Geoderma, 33: 135-155.

Valentin, C., 1981. Organisations Pelliculaires Superficielles de Quelques Sols de Région Subdésertique (Agadez, Rép. du Niger). Dynamique de formation et conskquences sur l'économie en eau. Ph.D. Thesis, Univ. Paris VII, Paris, 213 pp.

*Valentin, C., 1986. Surface crusting of arid sandy soils. In: F. Callebaut, D. Gabriels and M. De Boodt (Editors), Assessment of Soil Surface Sealing and Crusting. Flanders Research Center for Soil Erosion and Soil Conservation, Ghent, pp. 40-47.

*Valentin, C., 1991. Surface crusting in two alluvial soils of northern Niger. Geoderma, 48: 201-222.

25-29.

--1


Valentin, C. and Ruiz Figueroa, J.F., 1987. Effects of kinetic energy and water application rate on the development of crusts in a fme sandy loam soil using sprinkling irrigation and rainfall simulation. h. N. Fedoroff, L.M.-Bresson and M.A. Coutry (Editors), Soil

omorphology. Proc. VI1 Int. Working Meeting of Soil Micromorphology, Paris, July . Association Française pour 1'Etude du Sol, Plaisir, France, pp. 401-408.

in, C. and Bresson, L.M., 1992. Morphology, genesis and classification of surface crusts in loamy and sandy soils. Geoderma, 55: 225-245.

dentin, C. and Casenave, A., 1992. Infiltration into sealed soils as influenced by gravel cover. Soil Sci. Soc. Am. J., 56: 1167-1173. an der Watt, H.v.H. and Claassens, A.S., 1990. Effect of surface treatments on soil crusting and infiltration. Soil Technology, 3: 241-251.

*West, L.T., Bradford, J.M. and Norton, L.D., 1990. Crust morphology and hfïïtrability in surface soils from the southeast and midwest U.S.A. In: L.A. Douglas (Editor), Soil Micromorphology: A Basic and Applied Science. Proc. VI11 Int. Working Meeting on Soil Mircromorphology, San Antonio, Texas, July 1988. Developments in Soil Science 19, Elsevier, Amsterdam, pp. 107-113.

War, L.T., Chiang, S.C. and Norton, L.D., 1992. The morphology of surface crusts. In: M.E. Sumner and B.A. Stewart, Soil Crusting: chemical and Physical Processes, Proc. Ist Int. Symposium on Soil Crusting. Lewis Publishers, Boca Raton, pp. 73-92.

763

Micromorphological and hydraulic properties of surface crusts formed on a red earth soil in the semi-arid rangelands of eastern

Austraiia

R.S.B. Greene1* and A.J. Ringrose-Voas9 'CSIRO, Division of Wildlife and Ecology, PO Box 84, Lyneham, ACT2602, Australia.

2CSIR0, Division of Soils, GPO Box 639, Canberra, ACT 2601, Australia.

ABSTRACT

Greene, R.S.B. and Ringrose-Voase, A.J., 1994. Micromorphological and hydraulic properties of surface crusts formed on a red earth soil in the semi-arid rangelands of eastern Australia. In: A.J. Ringrose-Voase and G.S. Humphreys (Editors), Soil Micromorphology: Studies in Management and Genesis. Proc. IX Int. Working Mceting on Soil Micromorphology, Townsville, Australia, July 1992. Developments in Soil Science 22, Elsevier, Amsterdam, pp. 763-776.

Measurements were made of the microiorphological and hydraulic properties of surface crusts occurring on a Xerollic Haplargid soil (massive red earth) in the semi-arid rangelands of eastem Australia. The crusts were formed on hard, stony, runoff areas in a semi-arid mulga woodland on soil surfaces devoid of any plant or litter cover. The following treatments were used: (i) natural surface during a dry period, (i) ponded infiltration using a disc permeameter, (Ei) simulated rainfall, and (iv) simulated rainfall with protection against raindrop impact.

Micromorphological examination of the surface crusts distinguished four main categories of surfaces in each treatment, i.e. a matric crust, skelic crust, porphyric crust and a disturbed crust. The crusts were shown to be spatially variable and dynamic, with the proportion of the four categories related to the amount of wetting, raindrop impact, and surface flow received by the surfaces during each treatment. Wetting settles the disturbed category; impact creates skelic crust from some of the porphyric crust material and flow breaks up and mixes some of the skelic crust.

As these processes reduce the infiltration rate of the surface, the structural properties of these crusts have a major effect on surface hydrology and hence on vegetation distribution in semi-arid rangelands.

INTRODUCTION

The infiltration of rainfall and redistribution of runoff are critical in determining the long term stability of rangelands. They affect the subsequent spatial variation in available soil-water and hence have significant effects on diversity and production in the rangelands (Noy-Meir, 1973). Excessive runoff enhances soil-erosion hazards on sloping land and can cause off-site water and sediment damage downslope (Römkens et al.', 1990).

* Present address: Department of Geography, Australian National University, Canberra, ACT 0200, Australia

soil surface crust formation : contribution of...

Documents